Based on reporting by Forbes.
Google has introduced a new privacy-focused control in Chrome that allows users to delete locally stored AI model data from their device. The change affects Chrome features that rely on on-device artificial intelligence rather than cloud processing.
According to Google, some Chrome features such as scam detection, smart text analysis, and security warnings rely on local AI models stored directly on the user’s device. Until now, there was no clear option to remove this data once it had been downloaded.
The new toggle, available in Chrome’s privacy and security settings, allows users to delete this on-device AI data at any time. Once removed, the data can be re-downloaded automatically if the user continues using AI-powered features.
While Google emphasizes that the AI data never leaves the device, privacy advocates note that giving users explicit control over local AI storage is an important step toward transparency. The update comes amid growing scrutiny of how browsers handle AI-driven features.
Source: Forbes